how to install Deep Learning Toolbox in MATLAB Runtime

Question

bo 2023-4-19

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1949593-how-to-install-deep-learning-toolbox-in-matlab-runtime

评论： Steven Lord 2023-4-20

There is no option to select toolboxes when installing MATLAB Runtime R2022a (9.12). The Deep Learning Toolbox is not installed by default.

Is there a way to install the Deep Learning Toolbox in Runtime via modify the initial file or configuration?

Thanks.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Walter Roberson 2023-4-19

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1949593-how-to-install-deep-learning-toolbox-in-matlab-runtime#answer_1218623

No. Toolboxes cannot be installed in MATLAB Runtime.

In particular the only parts of Deep Learning Toolbox that can be deployed are the parts having to do with classifying or predicting data based upon pre-trained networks. Nothing having to do with training networks can be deployed.

2 个评论
显示无隐藏无

bo 2023-4-19

I find partial toolboxes are installed in Runtime.

Walter Roberson 2023-4-19

None the less, there is no control over what gets put in there.

请先登录，再进行评论。

Answer 2

Steven Lord 2023-4-19

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1949593-how-to-install-deep-learning-toolbox-in-matlab-runtime#answer_1218928

The MATLAB Runtime is not a way to run arbitrary MATLAB code for free. It is used to run applications created by MATLAB Compiler, and will only run code that was present when the application was built.

From the documentation "MATLAB Runtime only works on MATLAB code that was encrypted when the deployable archive was built. Any function or process that dynamically generates new MATLAB code will not work against MATLAB Runtime."

3 个评论
显示 1更早的评论隐藏 1更早的评论

Walter Roberson 2023-4-20

在 MATLAB Online 中打开

@bo

In one set of code, build your network and train it and get out a trained network. Save that network to a .mat file.

In another program, load() the network. You will probably need to use %#function naming the same class() that the network is, such as

%#function SeriesNetwork
datastruct = load('MyNetwork.mat', 'net');
net = datastruct.net;

Once the network is loaded, you can use it to predict().

You can use MATLAB Compiler on the code that does the load() and predict(). When you do that, MATLAB Compiler will automatically bundle in any necessary run-time functions as part of the CTF archive.